One of the strangest things about the analytics movement, in retrospect, was the number of celebrities it generated. I don’t mean because spreadsheet-wielding math geeks make for unlikely celebrities; in a world where Mark Zuckerberg has a prestige biopic and Julius Erving does not, that culture-hero ship left port a long time ago. I mean because the movement itself was explicitly opposed to the kind of soft, reputation-based cultural authority that celebrity represents. That kind of authority was supposed to belong to the old guard, the grizzled scouts who bragged about trusting their guts and the TV commentators who yammered on about clutchness and momentum and sticktoitiveness. It was supposed to belong to the GMs who overpaid for aging superstars in the twilights of their careers. It was supposed to belong, for that matter, to those creaky superstars. The new guard, the business-school dropouts wearing sloppy khakis and toting laptop bags, was supposed to value objectivity, not influence. Pure reason doesn’t hang out with Brad Pitt.
But even before Pitt played Billy Beane, the already-famous Oakland A’s GM and hero of Michael Lewis’s massively bestselling Moneyball, analytics already had created more household names than it had deposed. There was Bill James, the father of sabermetrics; by 2011, the year the film version of Moneyball came out, Theo Epstein had won two World Series championships with the Red Sox, and Nate Silver, having made the leap from analyzing baseball stats to analyzing political polling, was already being discussed in hushed tones by the cognoscenti, as if what he did was magic.
It wasn’t magic, as the new analytics stars themselves would have been the first to insist: the models, or at least the principles that went into making them, were available to anyone; the analysts weren’t saying “trust my special and unique powers,” they were saying “trust the math.” As far as the general public was concerned, though—a public sold more on Lewis’s gripping narrative skills than on an independent audit of the utility of VoRP—special and unique powers was precisely what they seemed to possess.
The first two decades of the 21st century were full of examples of men—think of Steve Jobs—who acquired a kind of paradoxically oracular authority as the leaders of hyper-rationalist fields. Something similar happened in sports. To many outside observers, “the data” became a type of secret knowledge, not really all that different from the “scout’s intuition” and “experience” that preceded it. It was understood to be more scientific; the narratives it generated were maybe somewhat less accessible, but this was a small price to pay, because it was newer and (presumably) better. And so the interpreters of the data were able to acquire a notoriety and an authority that were based less on rigorous understanding of their methods and more on the perceived ingenuity and prestige of their new code.
On the field, analytics and its various innovations were wildly successful without quite overturning our core understanding of sports. Teams that originated analytics-based innovations—Beane’s As, with their obsession with market inefficiencies and on-base percentage; Daryl Morey’s Rockets, with their obsession with 3-point shooting and free throws—tended to outperform expectations but not win championships. Later, teams deploying the same principles, but with more expensive and/or star-driven rosters (the Red Sox, the Golden State Warriors), would win titles. Advanced statistical analysis transformed the way we think about sports on many levels; on the other hand, there is no market inefficiency quite like having LeBron James.
This week, two of the biggest stars of the analytics movement abruptly walked away from their sports. Beane will reportedly leave his job as executive vice president with the A’s to join a business group investing in European soccer. Morey resigned as GM of the Rockets to spend more time with his college-age kids, currently at home because of the pandemic, and to, per the vague report that appeared on ESPN.com, “explore what else might interest him.” Morey, who’d been with the Rockets since 2006 and GM since 2007, reportedly plans to return to basketball; Beane, who’d been with the A’s in some capacity since 1990, and who has been passionate about soccer for many years, seems unlikely to return to baseball. I don’t know whether there’s an advanced statistical model that could calculate the odds that we’d lose Moneyball and Moreyball at exactly the same moment. I imagine they would be fairly low, but then, improbable things happen all the time; that is why it is worth watching sports.
In any case, as Beane and Morey exit their respective arbitrage industries, it’s a good moment to look back on the analytics movement as a whole. It’s obviously the case that sports, the action on the field as well as the construction of teams and contracts, look very different now compared to the pre-Moneyball era. The influence of the movement Beane and Morey helped to head largely restructured the way teams do business and the way fans think about games. Its insights were and are real. Looking back, though, what’s more interesting to me is how the analytics movement itself was shaped by the cultural narratives of its moment.
Think about how neatly it dovetailed, for instance, with the other sorts of Michael Lewis-y stories that dominated a certain section of the cultural imagination in the 2000s: investment banking, the poker phenomenon, the rise of the tech sector and the figure of the coder, all those stories about obsessive, detail-oriented, logical, ruthless, aggressive, fragile, clinical young men who solved games, seized opportunities, and toppled inefficient sentimentalities. The logic of disruption was also the logic of the stats revolution: We’re smarter, let the chips fall where they may. Of course, sports front offices aren’t in a position to do damage in the way that, say, Facebook is. Still, it’s no coincidence that the sports nerd-hero emerged as a celebrity just as the rest of the culture—comic book movies, the glorification of the hoodie-wearing tech billionaire, etc.—was devoting itself to flattering the egos of young, mostly white, smart, laptop-toting men. Especially the ones who retained the fixations (superheroes, sports, computers) they’d had in adolescence.
It’s not a stretch, in other words, to say that part of the reason Moneyball was such a monster hit was that it represented a massive transfer of power from the jocks to the nerds, in the jocks’ own house. It empowered a generation of quick-thinking sports fans to see themselves as essentially above the games they followed, to believe that they were capable of mastering them at a higher level even than the athletes. Athletes, in fact, could now be reimagined as kind of secondary to the proceedings. They were game pieces, pawns to be moved into position by the real masters of sports: the GMs, who turned real life into a fantasy league.
Which I think also helps to explain another aspect of the analytics movement that has always confused me: the fact that it thinks of itself as a movement at all. Doesn’t “movement” seem a little lofty for something so granular, practical, and technical? So deeply skeptical of abstract conventions? But from the beginning, you could feel that paradox in the way people talked about Moneyball: It was on the one hand a commitment to unemotional, lowercase-e empiricism, but on the other hand it had this glow of higher truth. Like tech culture in its early days (like tech culture now, for that matter, but in its early days it seemed less delusional), it seemed to point to something larger, some more fundamental shift in outlook, than the relatively modest points it was specifically addressing. The comparative efficiency of bunting or whether people could order pet food on their phones don’t sound like conversations with messianic undercurrents. But when the culture is flooded with narratives celebrating super-gifted savants and their upending of all prior assumptions, that’s what you get. This is, of course, another sort of irrational sentimentality, but then, it’s very rare to come across a skeptical movement that isn’t basically sentimental about itself.
In the end, what might be most striking about Beane and Morey’s departures is how relatively minor they feel. At one time, losing the leading visionaries of analytics in baseball and basketball might have seemed to call into question the future of the whole approach. By this point, though, analytics has been so thoroughly assimilated that not even losing the biggest celebrities of the movement can really threaten its place. If the test of a successful revolutionary movement is often whether it becomes mainstream, analytics has succeeded beyond all measure. It made sports teams better and smarter; at the same time, it didn’t eliminate the irrationality and arbitrariness that affect all human institutions, including itself. At some point, another new way of looking at sports will come along, one that will seem to supersede everything that went before it, including analytics. And afterward, we’ll look back at that new way of looking and say, well, it extended what we already knew. We’ll say, it was a product of its time. We’ll say, it changed everything, but not by all that much.